The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
Multimodal Generation Corpus
A durability-focused research corpus of AI-generated creative workflow media — node-graph session snapshots and their associated rendered assets (thumbnails, stills, and encoded previews) collected from public community galleries for generative content platforms.
Motivation
Media corpora for generative workflows are commonly stored on ephemeral CDNs with high link-rot. This corpus packages a snapshot of such assets in a sharded, integrity-checked, encrypted archival format suitable for durability/replication research and offline analysis of generative content characteristics (encoding formats, size distributions, thumbnail vs full-asset trade-offs).
Format
shards/*.data— encrypted (AES-256-CTR) tar archives. Each shard is self-describing after decryption: the final tar entry___manifest.jsonlists every contained object with byte size and SHA-256 digest.- Objects are content-addressed (
<kind>/<aa>/<16-hex>.<ext>) — no source identifiers, usernames, or original URLs are stored. - Shard-level integrity: each commit message carries the shard's encrypted byte size; the inner manifest carries per-object digests.
Access / decryption
The corpus is intended for archival replication research; the encryption key is retained by the dataset owner and shared with collaborators on request (the shards double as an integrity-verified transport format — anyone can verify shard transport integrity via the file digests even without decryption keys).
Statistics
Sharded incrementally as the collection grows; see the commit history for shard-level records.
- Downloads last month
- 599